[PATCH] drm/amdgpu/swsmu: don't bail early on hw_setup on resume

[Liu, Zhan]

[AMD Public Use]

> -----Original Message-----
> From: Alex Deucher <alexdeucher at gmail.com>
> Sent: 2021/March/26, Friday 5:01 PM
> To: Deucher, Alexander <Alexander.Deucher at amd.com>
> Cc: Liu, Zhan <Zhan.Liu at amd.com>; amd-gfx at lists.freedesktop.org
> Subject: Re: [PATCH] drm/amdgpu/swsmu: don't bail early on hw_setup on
> resume
>
> Looks like RN is immune due to the way it's is_dpm_running function is
> implemented.  Maybe something like this is a better solution:
> https://patchwork.freedesktop.org/patch/426293/
>
> Alex

Thanks Alex! Yes, I agree, this solution looks better to me. Since we've already done the same thing
for RN, it totally makes sense for us to make VG immune as well. And we can limit the influence
to VG only in this case.

Its self-explanatory that this change can get the issue fixed, but I still gave it a run on VG
and confirmed the issue is indeed get fixed here.

I will add my "Reviewed-by" under your new patch.

>
> On Fri, Mar 26, 2021 at 10:12 AM Deucher, Alexander
> <Alexander.Deucher at amd.com> wrote:
> >
> > [AMD Official Use Only - Internal Distribution Only]
> >
> >
> > Can someone double check this on RN/CZN with S3 and S0ix?
> >
> > Alex
> >
> > ________________________________
> > From: Liu, Zhan <Zhan.Liu at amd.com>
> > Sent: Friday, March 26, 2021 1:46 AM
> > To: Deucher, Alexander <Alexander.Deucher at amd.com>;
> > amd-gfx at lists.freedesktop.org <amd-gfx at lists.freedesktop.org>
> > Cc: Deucher, Alexander <Alexander.Deucher at amd.com>
> > Subject: RE: [PATCH] drm/amdgpu/swsmu: don't bail early on hw_setup on
> > resume
> >
> > [AMD Official Use Only - Internal Distribution Only]
> >
> > > -----Original Message-----
> > > From: amd-gfx <amd-gfx-bounces at lists.freedesktop.org> On Behalf Of
> > > Alex Deucher
> > > Sent: 2021/March/26, Friday 12:38 AM
> > > To: amd-gfx at lists.freedesktop.org
> > > Cc: Deucher, Alexander <Alexander.Deucher at amd.com>
> > > Subject: [PATCH] drm/amdgpu/swsmu: don't bail early on hw_setup on
> > > resume
> > >
> > > The SMU comes back up with DPM enabled by the sbios, but the driver
> > > still has to set up the SMU/driver mailbox, etc.
> > >
> > > Signed-off-by: Alex Deucher <alexander.deucher at amd.com>
> >
> > Reviewed-by: Zhan Liu <zhan.liu at amd.com>
> >
> > > ---
> > >  drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c | 2 +-
> > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
> > > b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
> > > index d4b804c7b986..462917d4d5e2 100644
> > > --- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
> > > +++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
> > > @@ -1102,7 +1102,7 @@ static int smu_smc_hw_setup(struct
> smu_context
> > > *smu)
> > >  uint32_t pcie_gen = 0, pcie_width = 0;  int ret = 0;
> > >
> > > -if (adev->in_suspend && smu_is_dpm_running(smu)) {
> > > +if (!smu->is_apu && adev->in_suspend &&
> > > smu_is_dpm_running(smu)) {
> > >  dev_info(adev->dev, "dpm has been enabled\n");
> > >  /* this is needed specifically */
> > >  if ((adev->asic_type >= CHIP_SIENNA_CICHLID) &&
> > > --
> > > 2.30.2
> > >
> > > _______________________________________________
> > > amd-gfx mailing list
> > > amd-gfx at lists.freedesktop.org
> > >
> > _______________________________________________
> > amd-gfx mailing list
> > amd-gfx at lists.freedesktop.org